Loading...
|
Please use this identifier to cite or link to this item:
http://140.128.103.80:8080/handle/310901/31493
|
| Title: | 建構智慧化疾病分類編碼輔助系統-以一般外科為例 |
| Other Titles: | The Development of Intelligent Support Systems for Coding International Classification of Diseases-The Case Study of General Surgery Medical Record |
| Authors: | 張維書
Chang,Wei-Shu |
| Contributors: | 謝宛霖
Hsieh,Wan-Lin
工業工程與經營資訊學系 |
| Keywords: | ICD 10 CM /PCS;文字探勘;關鍵字
ICD 10 CM/PCS;Text Mining,;Keyword |
| Date: | 2019 |
| Issue Date: | 2019-12-16T01:56:12Z (UTC)
|
| Abstract: | 自1995年全全民健康保險開辦至今，國際疾病分類代碼(International Classification of Disease)為各醫療院所向健保署申請醫療費用給付之重要依據，疾病分類代碼直接影響DRGs(Diagnosis- related Groups)歸屬及費用給付，疾病分類代碼的正確性和適當性越高，才能確保病患所做的醫療處置都能向健保署申請合理的給付，疾病分類師編列代碼在醫療院所扮演著相當重要角色，然而，疾病分類代碼的編排方式，具有其分類架構及分類邏輯，關鍵字的判定及查找正確的代碼，成為疾病分類師編碼的第一重要步驟。現今疾病與處置編碼的工作，已有電子工具書可輔助疾病分類師使用關鍵字協助編碼，然而，在使用此電子工具書前，仍須自行從主要診斷或主要處置描述中，判定其關鍵字。對於資淺的疾病分類師，仍需花費大量的時間與容易產生錯誤。本研究運用「文字探勘」協助建構主要診斷與主要處置之關鍵字判定系統。首先，整理歸納個案醫院中2016與2017年所有主要診斷與主要處置中，每個案例中協助判斷其關鍵字之判定字以及所對應之關鍵字，用以建立「主要診斷」及「主要處置」關鍵字判定系統之知識庫。此外，並整理出判定字搜尋的規則，作為關鍵字判定系統的設計邏輯。本研究以Python進行編碼，並以PyInstaller將此判定系統包裝成執行檔，透過2018年之病歷案例來驗證系統之準確性。本研究最後證實此「主要診斷與主要處置之關鍵字判定系統」，可以有效地降低疾病分類師進行「主要診斷」及「主要處置」的關鍵字判斷，主要診斷個案，資淺疾病分類師必須耗費1620秒，資深疾病分類師必須耗費930秒，使用關鍵字判定系統只須耗費450秒；而對於主要處置個案，資淺疾病分類師必須耗費2070秒，資深疾病分類師必須耗費1200秒，使用關鍵字判定系統只須耗費540秒。
Since the inception of the universal health Insurance in 1995, the International Classification code has become an important issue for medical institutions to apply to the health insurance Department for medical expenses. Disease classification code can directly affects DRGs attribution and expense payment. Disease classifiers play an important role on the medical institutions to make sure classification codes are organized based on their architecture and logic. Taiwan Government and some companies develop electronic tool book to assist disease classifiers while searching for correct disease codes. However, disease classifiers should be able to identify key words of diagnosis statement and treatment statement before using the electronic tool book. Therefore, junior disease classifiers still need to spend lots of time and it is easy to cause human errors. This study uses "text mining" to assist in the construction of keyword determination system for principal diagnosis and principal procedure. First of all,summary of case hospitals in 2016 and 2017 of all principal diagnoses and principal procedure, each case assists in judging the keyword's decision word and the corresponding keyword, knowledge base for establishing "principal diagnosis" and "principal procedure" keyword determination system. In addition,and sort out the rules for judging word search,as the design logic of keyword determination system. This study adopts Python for coding and PyInstaller for packing the system into an execution file. Then, the execution file is used to verify the accuracy of the system through a 2018-year medical record case.This study finally confirms that this "keyword determination system for principal diagnosis and principal procedure", can effectively reduce the disease classifier to carry out "principal diagnosis" and "principal procedure" keyword judgment,principal diagnostic case,disease division must take 1620 seconds, deep disease division must take 930 seconds, use the word to determine that the system takes only 450 seconds, and for the principal procedure, disease division must take 2070 seconds, deep disease division must take 1200 seconds, use the word to determine that the system takes only 540 seconds . |
| Appears in Collections: | [工業工程與經營資訊學系高階醫務工程與管理碩士在職專班] 碩士論文
|
Files in This Item:
| File |
Description |
Size | Format | |
|---|
| 107THU00030035-001.pdf | | 1422Kb | Adobe PDF | 29 | View/Open |
|
All items in THUIR are protected by copyright, with all rights reserved.
|
